Marginal Probabilities of the Extremes of ENSO Events for Temperature andPrecipitation in the Southeastern United States

摘要

The changes in seasonal average maximum and temperature and seasonal mean monthlyprecipitation during El Nino/ Southern Oscillation (ENSO) events are assessed at 36 stations in the southeastern United States by classifying 40 years (1948-1987) of monthly climate data as occurring during an El Nino (anomalously warm equatorial Pacific Ocean), El Viejo (anomalously cold equatorial Pacific Ocean), or neutral event using sea surface temperature data from the equatorial Pacific Ocean. The statistical distributions of the seasonal data are estimated by repeated sampling of available monthly data (a 'bootstrap technique'). The difference in means among the El Nino, El Viejo and neutral events are determined. The marginal probability is defined as the probability of a three-month seasonal average exceeding one standard deviation from the mean of all 40 years in the research period for a climate variable given an El Nino or El Viejo event is occurring. These are calculated for every station and each of ten different three-month seasons during an 'ENSO year', defined as running from October through the following September. The results indicate the southeastern United States tends to be cooler and wetter during El Nino events and warmer and drier during El Viego events. In contrast to previous studies, the magnitude of the cooling does not seem to be meaningful for El Nino events. Thus, the cold ENSO extreme called El Viejo seems to be most important for the extreme southeastern states of Florida, Georgia and South Carolina.